What Photographers Ought to Know About Bit Depth. 8 Bit vs 16 Bit Images
In the world of photography, we encounter a number of debates among the experts and amateurs: Film versus digital, RAW versus JPG, Canon versus Nikon… Oh goodness, let’s not forgot Mac versus Windows, though Mac is better and I am glad you agree! Wait, one more. How about 8 bit vs 16 bit images?
Have you ever wondered about the bit depth? Ever asked yourself whether you should work with 8 bit or 16 bit images? Perhaps you’ve worked with 16 bit images without knowing the full benefits. Anyway, if you never knew the difference, it is time to make things clear.
But First, a Bit on Pixel Identity
Let’s start with the fun geeky stuff. We are going to identify a pixel. A pixel is the smallest component of an image on a screen, usually a colored dot. The greater the number of pixels per inch the greater the resolution. Each pixel is assigned a numeric value. In the very beginning, computers assigned a value of either 1 or 0 to each pixel, which resulted with the image being black or white. This was known as a 1 bit image.
As the technology progressed, 2 bit images came along with such values as 00, 01, 10, 11. This endowed images with four different shades. Amazing, right? Not quite. 8 bits later bring us to more modern times, and that’s when things get shady. Get it?
Remember Photoshop Channels?
In the world of digital imaging, each pixel is created through a combination of three primary colors: red, green, blue, with each color referred as a color channel. Have you ever noticed the three channels in Photoshop, located under the channel panel? Well, each channel has intensity values specified by its bit depth. The depth for each primary color is known as bits per channel.
You might hear some photographers and designers use the term bits per pixel. They are simply referring to the sum of these three colors, so don’t get confused. An 8 bit binary digit equals to 256 shades between black and white. With each channel using 256 tones, we can create a 24 bit image with, potential, 16.7 billion colors (r=256xg=256xb=256).
That’s a lot of color, considering we can only see about 10 million. But this is barely enough knowing that with constant image enhancement, we kill hundreds of thousands of these pixels. Yes, we are participating in a pixel-genocide, and we don’t even know it! But the Holy Curve is watching and here is no way to hide from the histogram. So beware. For this reason, I suggest that we work with 16 bit images when necessary, 32 bit with HDR, but I won’t go there. Not yet!
|Bits Per Pixel||Number of Colors|
Goodness, I am confused!
I feel your pain, since I am having a hard time explaining this, so that it can make sense. Let’s get straight to the point. Look at the table above. Notice that the 8 bit files have 256 colors while the 16 bit files have 65,536 colors. One is more flexible than the other. It just depends on how you’re going to use it, and whether you see the benefits of choosing 16 bit.
Some DSLR’s provide 12, 14 and 16 bit raw images, therefore, it is wise to take advantage of this option. Say we are about to work on a black and white image, provided it’s a RAW file. The first thing I’d do, in Camera Raw, is change the Depth to 16 bits/channel.
The file will be larger, since I’ll be working with 65,536 shades, instead of 256. Just imagine the lovely tonality between the dark shadows and bright highlights! Yes, I can hear you asking, “What if I shoot jpegs and convert them to 16 bit in Photoshop?” Sorry, it doesn’t work that way. Keep reading, and find out which bit you might bite on.
Finally! 8 Bit vs 16 Bit Images
8 Bit Images
8 Bit images have 256 colors per channel with over 16 trillion possible tones. Enhancing images in this mode increase the chances of banding since there is less data to work with. Banding occurs when there is a lack of tonal values, creating posterization, which is not a pleasing effect if there are smooth gradients in a picture, whether it be color or black and white. This unwanted artifact often shows up in pictures with smooth skies or plain backdrops. If you know your prints will be small and there is very little art work to be applied, then 8 bit images are just fine. Perhaps it is safe to say that it all depends on the image subject.
16 Bit Images
16 bit images have 16 levels of intensity, with over 280 trillion possible tones. If you are planning on printing large black and white prints, I suggest that, from Camera Raw, you select 16 bits/channel from the Depth drop-down menu. The files will be larger, but you will have greater tonality range which prevents aliasing, also known as banding. For printing, we need to convert our 16 bit to 8 bit files. Further image enhancing should end upon the mentioned conversion. Reversal is not beneficial.
A Bit on Gamut
Photographers who use large gamut color spaces, such as Adobe RGB (1998) and PhotoPro RGB, should work with larger bit depth. The reason is obvious. The larger the color space the more bits are needed. Imagine painting a room. The paint is a bit depth and the wall is a color space. The bigger the wall, the more paint you’ll need, otherwise you’ll have white patches. I’ll stop here, since this topic can get quite saturated in technical detail.
How do I get 16 bit images?
To get 16 bit images, we have to shoot RAW files. Prior to opening them in Photoshop, ensure that the 16 bit/channel field has been selected. In Photoshop Camera Raw, it is located under the Depth drop-down menu. Lets keep in mind that converting 8 bit jpg images to 16 bit in Photoshop offers no benefits.
Don’t go back and forth
Not all printers, including those used at D&M Imaging, can print 16 bit images. They are quite rare. If you stick with 16 bit, enhance your work in 16 bit. Don’t convert to 8 bit and then apply more work. Also, don’t convert to 8 bit, and then back to 16 bit, in order to make further adjustments such as color, retouching, etc… There are absolutely no benefits of doing that.
Bit depth matters in post production. When we make large prints, create HDR photography, or care to preserve great range of tonal values, it is wise to work on our images in a 16 bit mode, which gives us greater tonality and color values. Once we know that no more work will be applied to our images, we can flatten the layers, convert the images back to 8 bit and send them to the printer. This conversion, from 16 bit to 8 bit, strips all the extra data, but this will not affect the quality of our final image.
We just have to make sure that our editing, a better word would be enhancing, should end with the conversion. I am sure that some photographers will protest, claiming they see no benefit in 16 bit images, but with each adjustment, we are altering pixel values which results in data loss. It is simple. More money buys us more property. The same applies to bit depth.
Notes + Tips
- In B&W, higher bit means greater tonal values.
- Bit depth indicates the color of each pixel.
- Bit depth has two aspects, gamut and color precision.
- We can only see 10 million colors.
- We cannot print 16 bit images.
- Use 16 bit files for HDR photography.
- Use 16 bit files in B&W photography.
- With HDR, I suggest we work with 32 bit files.
- Use 16 bit files in pictures with greater natural gradients, such as skies.
- For better results and bigger prints, work in 16 bit mode, when done, convert to 8 bit.